TCAP FAC‐WIN6 Elite Barley GWAS Panel QTL. I. Barley Stripe Rust Resistance QTL in Facultative and Winter Six‐Rowed Malt Barley Breeding Programs Identified via GWAS

Abstract

ABSTRACTThe fungal disease barley stripe rust (BSR, causal agent Puccinia striiformis f. sp. hordei) can greatly reduce yield and quality of malt barley (Hordeum vulgare L.), a valuable crop in the distilling and brewing industries. Yield losses of up to 70% have been reported. Resistant cultivars are the most sustainable defense against BSR, especially those cultivars with pyramids of quantitative disease resistance (QDR) genes. The Oregon State University (OSU) barley breeding program has selected for BSR resistance for decades and has released cultivars with BSR resistance. We wanted to determine the number and effect of the loci underlying that resistance, as well as whether it was conferred by QDR or race‐specific qualitative loci. To test these, we used BSR resistance quantitative trait loci (QTL) mapping via genomewide association studies (GWAS) in adult‐plant field trials (3 yr, one location) and race‐specific seedling greenhouse trials (five races). The FAC‐WIN6, a GWAS panel composed of 300 lines that represent the OSU and University of Minnesota facultative and winter six‐rowed malt barley breeding programs, was used in genotyping and phenotyping. The FAC‐WIN6 was developed as part of the USDA‐ARS Triticeae Coordinated Agricultural Project but was not developed specifically for BSR resistance. In total, we detected three race‐specific QTL in greenhouse trials (one of which was also significant in field trials) and 14 QTL significant in field trials only with 2.3 to 20.7% additive reduction in percentage disease severity. At least five QTL were considered novel. These results indicated that the OSU breeding program had both QDR and race‐specific loci for BSR resistance. We suspected that its BSR resistance distributions in the FAC‐WIN6 might align with the major population structure patterns in ways that would be difficult to account for with a GWAS model alone. Therefore, based on post hoc diagnostics, we report a qualitative “usability” level for each QTL.